the-other-tim-brown commented on code in PR #13422:
URL: https://github.com/apache/hudi/pull/13422#discussion_r2141396107
##########
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/client/SparkRDDWriteClient.java:
##########
@@ -391,6 +391,11 @@ public void releaseResources(String instantTime) {
SparkReleaseResources.releaseCachedData(context, config, basePath,
instantTime);
}
+ @Override
+ public SparkRDDWriteClient<T> createNewClient() {
+ return new SparkRDDWriteClient<>(context,
HoodieWriteConfig.newBuilder().withProperties(config.getProps()).build());
Review Comment:
That is the intention here. You cannot reuse timeline servers between
clients running in parallel. The view of the table is modified and you have an
inconsistent/non-deterministic execution of phases relying on the "latest" base
file or file slice.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
